# Dense Passage Retrieval for Open-Domain Question Answering

Vladimir Karpukhin, Barlas Oğuz, Sewon Min, Patrick Lewis, Ledell Wu, Sergey Edunov, Danqi Chen, Wen-tau Yih

2020 | arXiv:2004.04906

Source: https://arxiv.org/abs/2004.04906

Reading scope: abstract-and-metadata

Editorial update: 2026-10-20

A dual encoder retrieves candidate passages using learned dense representations.

## Why this matters

Retrieval quality can be studied separately from the answering model.

## The proposal

DPR examines whether learned dense representations can provide practical passage retrieval for open-domain QA, using a comparatively simple dual-encoder framework.

## What was inspected

Its abstract reports stronger retrieval than a Lucene BM25 baseline on evaluated datasets. That is an author-reported comparison, not a claim that every dense index beats lexical search.

## Editorial interpretation

Before evaluating an answer, inspect whether the needed passage entered the candidate set. Retrieval errors and generation errors call for different interventions.

## Mechanism

Encode questions and passages.

Search the shared representation space.

Pass candidate contexts to a QA system.

## Limits

Dense retrieval superiority is not universal.

Dataset and negative-sampling details remain unreviewed here.